Prevention of Significant Deterioration (PSD) and Nonattainment New Source Review (NSR): Debottlenecking, Aggregation, and Project Netting, 54235-54252 [E6-15248]

Agencies

• ENVIRONMENTAL PROTECTION AGENCY

[Federal Register Volume 71, Number 178 (Thursday, September 14, 2006)]
[Proposed Rules]
[Pages 54235-54252]
From the Federal Register Online via the Government Printing Office [www.gpo.gov]
[FR Doc No: E6-15248]


========================================================================
Proposed Rules
                                                Federal Register
________________________________________________________________________

This section of the FEDERAL REGISTER contains notices to the public of 
the proposed issuance of rules and regulations. The purpose of these 
notices is to give interested persons an opportunity to participate in 
the rule making prior to the adoption of the final rules.

========================================================================


Federal Register / Vol. 71, No. 178 / Thursday, September 14, 2006 / 
Proposed Rules

[[Page 54235]]



ENVIRONMENTAL PROTECTION AGENCY

40 CFR Parts 51 and 52

[EPA-HQ-OAR-2003-0064; FRL-8219-6]
RIN 2060-AL75


Prevention of Significant Deterioration (PSD) and Nonattainment 
New Source Review (NSR): Debottlenecking, Aggregation, and Project 
Netting

AGENCY: Environmental Protection Agency (EPA).

ACTION: Proposed rule.

-----------------------------------------------------------------------

SUMMARY: The EPA proposes revisions to the regulations governing the 
major NSR programs mandated by parts C and D of title I of the Clean 
Air Act (CAA). These proposed changes reflect EPA's consideration of 
the Agency's 2002 Report to the President and its associated 
recommendations as well as discussions with various stakeholders 
including representatives of environmental groups, State and local 
governments, and industry. We propose to change how emissions from 
emissions units upstream or downstream from the unit(s) undergoing a 
physical change or change in the method of operation are included in 
the calculation of an emissions increase for the project. Also, these 
proposed changes would clarify and codify our policy of when emissions 
increases from multiple projects are to be aggregated together to 
determine NSR applicability. Finally, we are clarifying how emissions 
decreases from a project may be included in the calculation to 
determine if a significant emissions increase will result from a 
project. We intend the proposed rules to improve implementation of the 
program by articulating and codifying principles for determining major 
NSR applicability that we currently address through guidance only.
    We are seeking comment on all aspects of this proposed rule. This 
proposal seeks public comment in accordance with section 307(d) of the 
CAA and should not be used or cited in any litigation as a final 
position of the Agency.

DATES: Comments. Comments must be received on or before November 13, 
2006. Under the Paperwork Reduction Act, comments on the information 
collection provisions must be received by OMB on or before October 16, 
2006.
    Public Hearing. If anyone contacts EPA requesting a public hearing 
by September 28, 2006, we will hold a public hearing approximately 30 
days after publication in the Federal Register.

ADDRESSES: Submit your comments, identified by Docket ID No. EPA-HQ-
OAR-2003-0064 by one of the following methods:
     www.regulations.gov: Follow the online instructions for 
submitting comments.
     E-mail: a-and-r-docket@epa.gov.
     Mail: Air and Radiation Docket and Information Center, 
Environmental Protection Agency, Mailcode: 6102T, 1200 Pennsylvania 
Ave., NW., Washington, DC 20460. In addition, please mail a copy of 
your comments on the information collection provisions to the Office of 
Information and Regulatory Affairs, Office of Management and Budget 
(OMB), Attn: Desk Officer for EPA, 725 17th St., NW., Washington, DC 
20503.
     Hand Delivery: Environmental Protection Agency, EPA West 
Building, Room B102, 1301 Constitution Ave., NW., Washington, DC. Such 
deliveries are only accepted during the Docket's normal hours of 
operation, and special arrangements should be made for deliveries of 
boxed information.
    Instructions: Direct your comments to Docket ID No. EPA-HQ-OAR-
2003-0064. EPA's policy is that all comments received will be included 
in the public docket without change and may be made available online at 
www.regulations.gov, including any personal information provided, 
unless the comment includes information claimed to be Confidential 
Business Information (CBI) or other information whose disclosure is 
restricted by statute. Do not submit information that you consider to 
be CBI or otherwise protected through www.regulations.gov or e-mail. 
The www.regulations.gov Web site is an ``anonymous access'' system, 
which means EPA will not know your identity or contact information 
unless you provide it in the body of your comment. If you send an e-
mail comment directly to EPA without going through www.regulations.gov, 
your e-mail address will be automatically captured and included as part 
of the comment that is placed in the public docket and made available 
on the Internet. If you submit an electronic comment, EPA recommends 
that you include your name and other contact information in the body of 
your comment and with any disk or CD ROM you submit. If EPA cannot read 
your comment due to technical difficulties and cannot contact you for 
clarification, EPA may not be able to consider your comment. Electronic 
files should avoid the use of special characters, avoid any form of 
encryption, and be free of any defects or viruses. For additional 
information about EPA's public docket, visit the EPA Docket Center 
homepage at https://www.epa.gov/epahome/dockets.htm. For additional 
instructions on submitting comments, go to section I.B of the 
Supplementary Information section of this document.
    Docket: All documents in the docket are listed in the 
www.regulations.gov index. Although listed in the index, some 
information is not publicly available, e.g., CBI or other information 
whose disclosure is restricted by statute. Certain other material, such 
as copyrighted material, is not placed on the Internet and will be 
publicly available only in hard copy form. Publicly available docket 
materials are available either electronically in www.regulations.gov or 
in hard copy at the Air and Radiation Docket and Information Center, 
EPA/DC, EPA West Building, Room B102, 1301 Constitution Ave., NW., 
Washington, DC. The Public Reading Room is open from 8:30 a.m. to 4:30 
p.m., Monday through Friday, excluding legal holidays. The telephone 
number for the Public Reading Room is (202) 566-1744, and the telephone 
number for the Air and Radiation Docket and Information Center is (202) 
566-1742.

    Note: The EPA Docket Center suffered damage due to flooding 
during the last week of June 2006. The Docket Center is continuing 
to operate. However, during the cleanup, there will be temporary 
changes to Docket Center telephone numbers, addresses, and hours of 
operation for people who wish to make hand deliveries or visit the 
Public Reading Room to view documents. Consult EPA's Federal 
Register notice at 71 FR 38147

[[Page 54236]]

(July 5, 2006) or the EPA Web site at www.epa.gov/epahome/
dockets.htm for current information on docket operations, locations 
and telephone numbers. The Docket Center's mailing address for U.S. 
mail and the procedure for submitting comments to 
www.regulations.gov are not affected by the flooding and will remain 
the same.


FOR FURTHER INFORMATION CONTACT: Mr. David J. Svendsgaard, Air Quality 
Policy Division, Office of Air Quality Planning and Standards (C504-
03), U.S. Environmental Protection Agency, Research Triangle Park, NC 
27711, telephone (919) 541-2380, fax number (919) 541-5509, e-mail 
address svendsgaard.dave@epa.gov.

SUPPLEMENTARY INFORMATION:

I. General Information

A. Does This Action Apply to Me?

    Entities affected by this rule include sources in all industry 
groups. The majority of sources potentially affected are expected to be 
in the following groups:

------------------------------------------------------------------------
           Industry Group            SIC \a\           NAICS \b\
------------------------------------------------------------------------
Electric Services..................      491  221111, 221112, 221113,
                                               221119, 221121, 221122
Petroleum Refining.................      291  324110
Industrial Inorganic Chemicals.....      281  325181, 325120, 325131,
                                               325182, 211112, 325998,
                                               331311, 325188
Industrial Organic Chemicals.......      286  325110, 325132, 325192,
                                               325188, 325193, 325120,
                                               325199
Miscellaneous Chemical Products....      289  325520, 325920, 325910,
                                               325182, 325510
Natural Gas Liquids................      132  211112
Natural Gas Transport..............      492  486210, 221210
Pulp and Paper Mills...............      261  322110, 322121, 322122,
                                               322130
Paper Mills........................      262  322121, 322122
Automobile Manufacturing...........      371  336111, 336112, 336211,
                                               336992, 336322, 336312,
                                               336330, 336340, 336350,
                                               336399, 336212, 336213
Pharmaceuticals....................      283  325411, 325412, 325413,
                                               325414
------------------------------------------------------------------------
\a\ Standard Industrial Classification.
\b\ North American Industry Classification System.

    Entities affected by the rule also include States, local permitting 
authorities, and Indian tribes whose lands contain new and modified 
major stationary sources.

B. Where Can I Get a Copy of This Document and Other Related 
Information?

    In addition to being available in the docket, an electronic copy of 
this proposal will also be available on the WWW. Following signature, a 
copy of this notice will be posted in the regulations and standards 
section of our NSR home page located at https://www.epa.gov.

C. What Should I Consider as I Prepare My Comments for EPA?

    1. Submitting CBI. Do not submit information that you consider to 
be CBI electronically through www.regulations.gov or e-mail. Clearly 
mark the part or all of the information that you claim to be CBI. For 
CBI information in a disk or CD ROM that you mail to EPA, mark the 
outside of the disk or CD ROM as CBI and then identify electronically 
within the disk or CD ROM the specific information that is claimed as 
CBI. In addition to one complete version of the comment that includes 
information claimed as CBI, a copy of the comment that does not contain 
the information claimed as CBI must be submitted for inclusion in the 
public docket. Information so marked will not be disclosed except in 
accordance with procedures set forth in 40 CFR part 2.
    2. Tips for Preparing Your Comments. When submitting comments, 
remember to:
     Identify the rulemaking by docket number and other 
identifying information (e.g., subject heading, Federal Register 
proposal publication date and reference page number(s)).
     Follow directions--The EPA may ask you to respond to 
specific questions or organize comments by referencing a Code of 
Federal Regulations (CFR) part or section number.
     Explain why you agree or disagree; suggest alternatives 
and provide substitute language for your requested changes.
     Describe any assumptions and provide any technical 
information and/or data that you used.
     If you estimate potential costs or burdens, explain how 
you arrived at your estimate in sufficient detail to allow for it to be 
reproduced.
     Provide specific examples to illustrate your concerns, and 
suggest alternatives.
     Explain your views as clearly as possible, avoiding the 
use of profanity or personal threats.
     Make sure to submit your comments by the specified comment 
period deadline.
    Commenters wishing to submit proprietary information for 
consideration must clearly distinguish such information from other 
comments and clearly label it as CBI. Send submissions containing such 
proprietary information directly to the following address, and not to 
the public docket, to ensure that proprietary information is not 
inadvertently placed in the docket: Attention: Mr. Roberto Morales, 
U.S. Environmental Protection Agency, OAQPS Document Control Officer, 
109 TW Alexander Drive, Room C404-02, Research Triangle Park, NC 27711. 
The EPA will disclose information identified as CBI only to the extent 
allowed by the procedures set forth in 40 CFR part 2. If no claim of 
confidentiality accompanies a submission when it is received by EPA, 
the information may be made available to the public without further 
notice to the commenter.

D. How Can I Find Information About a Possible Hearing?

    People interested in presenting oral testimony or inquiring as to 
whether a hearing is to be held should contact Ms. Pam Long, Air 
Quality Planning Division, Office of Air Quality Planning and Standards 
(C504-03), U.S. Environmental Protection Agency, Research Triangle 
Park, NC 27711, telephone (919) 541-0641, fax number (919) 541-5509, e-
mail address long.pam@epa.gov, at least 2 days in advance of the public 
hearing. People interested in attending the public hearing must also 
call Ms. Long to verify the time, date, and location of the hearing. 
The public hearing will provide interested parties the opportunity to 
present data, views, or arguments concerning the proposed action. If a 
public hearing is held, it will be held at 9 a.m. in EPA's Auditorium 
in Research

[[Page 54237]]

Triangle Park, North Carolina, or at an alternate site nearby.

E. How Is This Preamble Organized?

    The information presented in this preamble is organized as follows:

I. General Information
    A. Does This Action Apply to Me?
    B. Where Can I Get a Copy of This Document and Other Related 
Information?
    C. What Should I Consider as I Prepare My Comments for EPA?
    D. How Can I Find Information About a Possible Hearing?
    E. How Is This Preamble Organized?
II. Introduction
III. Debottlenecking
    A. Background
    B. Overview of This Proposed Action
    C. Discussion of Issues Under Proposed Debottlenecking Approach
IV. Aggregation
    A. Background
    B. Overview of This Proposed Action
    C. Discussion of Issues Under Proposed Aggregation Approach
V. Project Netting
    A. Background
    B. Overview of This Proposed Action
VI. Statutory and Executive Order Reviews
A. Executive Order 12866--Regulatory Planning and Review
B. Paperwork Reduction Act
C. Regulatory Flexibility Analysis
D. Unfunded Mandates Reform Act
E. Executive Order 13132--Federalism
F. Executive Order 13175--Consultation and Coordination With Indian 
Tribal Governments
G. Executive Order 13045--Protection of Children From Environmental 
Health Risks and Safety Risks
H. Executive Order 13211--Actions Concerning Regulations That 
Significantly Affect Energy Supply, Distribution, or Use
I. National Technology Transfer and Advancement Act
VII. Statutory Authority

II. Introduction

    In May 2001, President Bush's National Energy Policy Development 
Group issued findings and key recommendations for a National Energy 
Policy. This document included numerous recommendations for action, 
including a recommendation that the EPA Administrator, in consultation 
with the Secretary of Energy and other relevant agencies, review NSR 
regulations, including administrative interpretation and 
implementation.\1\ The recommendation requested that we issue a report 
to the President on the impact of the regulations on investment in new 
utility and refinery generation capacity, energy efficiency, and 
environmental protection.
---------------------------------------------------------------------------

    \1\ For an overview of the major NSR program, see 67 FR 80187-
80188.
---------------------------------------------------------------------------

    In response, in June 2001, we issued a background paper giving an 
overview of the NSR program. This paper is available on the Internet at 
https://www.epa.gov/nsr/publications.html. We solicited public comments 
on the background paper and other information relevant to the NSR 90-
day Review and Report to the President. During our review of the NSR 
program, we met with more than 100 groups, held four public meetings 
around the country, and received more than 130,000 written comments. 
Our Report to the President and our recommendations in response to the 
energy policy were issued on June 13, 2002. A copy of this information 
is available at https://www.epa.gov/nsr/publications.html.
    We have previously finalized responses to energy policy 
recommendations on December 31, 2002 (67 FR 80186) and October 27, 2003 
(68 FR 61248).\2\ These proposed regulations for ``aggregation'' and 
``debottlenecking'' are a further response to the remaining 
recommendations. We also are proposing a change to our past policy for 
project netting. We believe that these proposed rules would provide 
greater regulatory certainty while preserving the current level of 
environmental protection and benefit derived from the current NSR 
program.
---------------------------------------------------------------------------

    \2\ On March 17, 2006, the DC Circuit Court of Appeals vacated 
the October 27, 2003 rule. On June 30, 2006, the Court denied EPA's 
request for rehearing or, in the alternative, rehearing en banc with 
respect to this decision.
---------------------------------------------------------------------------

    This action proposes and requests comment on changes to the 
regulations for both the approval and promulgation of implementation 
plans and requirements for preparation, adoption, and submittal of 
implementation plans governing the NSR programs mandated by parts C and 
D of title I of the CAA. We also propose to include conforming changes 
to 40 CFR (Code of Federal Regulations) part 51, appendix S. This 
notice does not include specific regulatory language related to this 
section. Nonetheless, we intend to finalize these rule provisions in 
Appendix S, either at the time we finalize the remainder of these 
proposed revisions, or at the time that we finalize changes to 
incorporate the 2002 NSR improvements into Appendix S. We seek comment 
on incorporating these changes into Appendix S through this proposed 
rule, and will not seek additional comments before taking final action 
on the Appendix S changes.

III. Debottlenecking

A. Background

1. NSR Improvement Rule of 2002
    As noted above, EPA has already promulgated rules in response to 
the 2002 recommendations. On December 31, 2002, we finalized changes to 
NSR applicability for modifications at major stationary sources. 
Specifically, this rule promulgated changes for how to calculate 
emissions increases at sources that have undergone a physical change or 
change in the method of operation, or ``project.''
    As a result of the 2002 rules, the major NSR regulations now 
measure an emissions increase from a project by comparing the change in 
actual emissions before and after the change.\3\ Under this 
methodology, the actual annual emissions before the change are compared 
with the projected actual annual emissions after the change to 
determine if a physical or operational change would result in a 
significant increase in emissions. The major NSR regulations allow for 
consideration of an emissions unit's operating capacity in determining 
whether a change results in an emissions increase. Under the actual-to-
projected-actual test, a source can subtract from its post-project 
emissions those emissions that the unit ``could have accommodated'' 
during the baseline period and that are unrelated to the change 
(sometimes referred to as the ``demand growth exclusion''). That is, 
the source can emit up to its current maximum capacity without 
triggering major NSR under the actual-to-projected-actual test, as long 
as the increase is unrelated to the physical or operational change.
---------------------------------------------------------------------------

    \3\ Sources are allowed to use an actual-to-potential emissions 
test for NSR applicability that makes them not subject to reporting 
and recordkeeping requirements that are required under the new 
actual-to-projected-actual emissions test. See 67 FR 80197.
---------------------------------------------------------------------------

    Various governmental and nongovernmental entities sought judicial 
review of many aspects of the 2002 rules. In New York v. EPA, 413 F.3d 
3 (D.C. Cir. 2005) (``New York''), the Court largely upheld EPA's rules 
on projecting actual emissions resulting from a change. The Court held 
that the NSR modification requirement, which incorporates by reference 
CAA section 111(a)(4), ``unambiguously defines `increases' in terms of 
actual emissions.'' See New York, 413 F.3d at 39. The Court also upheld 
excluding from projections those increases attributable to ``demand 
growth.'' Id. at 33. Those emissions were increases that could have 
been accommodated by the

[[Page 54238]]

source prior to the change and which were unrelated to change.
    Most of the applicability test in the 2002 rule based emissions 
test on historical (actual) emissions; however, EPA also promulgated 
the Clean Unit exemption, which would have allowed a source to 
calculate its emissions increase based on its permitted emissions. 
While the Court upheld EPA on projected actual emissions, it vacated 
the method of calculating emissions for Clean Units. The Court held 
that EPA lacked authority to promulgate the Clean Unit provision, and 
in doing so, held that ``the plain language of the CAA indicates that 
Congress intended to apply NSR to changes that increase actual 
emissions instead of potential or allowable emissions.'' The Court held 
that the method for clean units would have impermissibly relied on a 
measure of emissions that was not based on actual emissions increases 
at the source.
2. What Is ``Debottlenecking''?
    A major stationary source often consists of multiple emitting and 
non-emitting units that comprise integrated processes at the source. As 
part of the operations of the source or within a process, various 
pieces of equipment may provide input to or accept output from other 
equipment or units at the source. These equipment and units at the 
source may have different operating capacities.
    When equipment and units of different capacities operate, one unit 
may constrain other units from operating at their full design capacity 
or maximum output rating either by limiting inputs to those other units 
or by limiting usable output. Such constraining equipment and units are 
commonly called ``bottlenecks'' in a process. The constrained emissions 
unit(s) can be situated in the process either in advance of the 
constraining emissions unit (i.e., ``upstream'') or after it 
(``downstream'').
    When a constraining unit or piece of equipment is changed to 
increase its capacity, another unit may increase its operations 
(depending on whether some or all of the constraint was removed) to 
provide input to the changed unit or use output from it. We have 
historically referred to this phenomenon as ``debottlenecking.'' This 
increased operation of the upstream or downstream emissions unit(s) can 
contribute to increased emissions from the unit(s).
    Our current regulations define a ``major modification'' as one in 
which a physical change or a change in the method of operation of a 
major stationary source results in a significant emissions increase of 
a regulated NSR pollutant and a significant net emissions increase of 
that pollutant at the source. See 40 CFR 52.21(b)(2). Based on this 
current regulation, the total increase in emissions that are included 
in determining if there will be a post-change significant emissions 
increase includes: (1) Increases occurring at all new or modified 
units, and (2) any other increases at existing emissions units not 
being modified that experience emissions increases as a result of the 
change.\4\ Under our current and prior rules, we have presumed that 
increases in emissions at a debottlenecked unit are caused by the 
project and, thus, included in determining NSR applicability for the 
project.
---------------------------------------------------------------------------

    \4\ Note that, later in this preamble, we propose to include 
decreases (along with increases) from emissions units in calculating 
the emissions change that results from a project (i.e., the first 
step of the NSR applicability analysis).
---------------------------------------------------------------------------

    The EPA's recommendation to the President directed changes to our 
``debottlenecking'' rule provisions, and we recognize that there has 
been confusion over our past policies for calculating emissions from 
debottlenecked units and from units experiencing an ``increase in 
utilization.'' While we are not defining the term ``debottlenecked 
unit'' in this proposed rule, we intend for these provisions, when 
finalized, to apply to any unchanged unit at a source that increases 
its utilization following a change elsewhere at the source.
3. How Does EPA Currently Implement Major NSR for Debottlenecking 
Changes?
    As stated above, the emission calculation for a new project 
includes the emission increases from all the units involved in a 
project. Any new unit's emission increase that results from the project 
is equal to the unit's potential to emit, or ``PTE.'' See 40 CFR 
52.21(a)(2)(iv)(d). For existing units, the emission increase 
associated with the project is based on the ``actual-to-projected-
actual'' test, and, under the current test, it includes increases not 
only from the unit(s) undergoing the change but also increases at any 
other unit at the major stationary source that are related to the 
change.\5\ In the past, EPA has generally assumed that emissions from 
debottlenecked units result from the proposed project.
---------------------------------------------------------------------------

    \5\ These emissions increase test requirements apply to sources 
in delegated jurisdictions. Some SIP-approved jurisdictions have not 
yet adopted EPA's rules into their SIP's, meaning that their 
previous rules apply for their sources until they adopt the 2002 
rules.
---------------------------------------------------------------------------

    Under the ``actual-to-projected-actual'' test, pre-change emissions 
are determined using the procedures for ``baseline actual emissions.'' 
As evident in 40 CFR 52.21(b)(48), different rules apply for 
determining baseline actual emissions depending on whether or not the 
source is an electric utility. Except for electric utility steam 
generating units, the major stationary source requesting the 
modification may use any consecutive 24-month period in the past 10 
years to determine the baseline actual emissions for the emissions 
unit(s) involved. This 10-year ``look back'' period is limited to 5 
years for electric utilities, but a different 24-month period outside 
of the 5-year window can be used if it is more representative of normal 
source operation. Post-change emissions are generally projected using 
the emissions unit's maximum annual rate, in tons per year, at which it 
is expected to emit a regulated NSR pollutant within five years 
following a change, less any amount of emissions that the unit could 
have accommodated during the selected 24-month baseline period and that 
are unrelated to the change. This final ``projected actual'' value, in 
tons per year, is the value you compare to the ``baseline actual 
emissions'' in order to determine, by summing the increases at various 
emissions units, whether the proposed project will result in a 
``significant'' emissions increase, as defined in the first step of the 
calculation. See 40 CFR 52.21(b)(23).\6\
---------------------------------------------------------------------------

    \6\ The EPA is developing a rule for electric generating units 
(EGU) that would change the test for net emissions increase for 
those units. See 70 FR 61081 (October 20, 2005).
---------------------------------------------------------------------------

    The actual-to-projected-actual test in the 2002 rules for existing 
emissions units applies not only to the unit(s) undergoing the change 
but also to any other existing emissions unit(s) at the source that 
experiences a change in emissions related to the project. Thus, the 
current EPA rules permit emissions increases from debottlenecked units 
(and any other unit that increases its utilization as a result of the 
project) to be calculated using an ``actual-to-projected-actual'' 
test.\7\ We believe this represents a fair reading of our current 
regulatory text for ``projected actual emissions'' found at 40 CFR 
52.21(b)(41).\8\
---------------------------------------------------------------------------

    \7\ Note that EPA does not require that sources use projected 
actual emissions to calculate their emissions increases. If a source 
prefers, it can calculate its emissions increases by comparing its 
past actual emissions to its future potential to emit.
    \8\ We note that some confusion was caused by a footnote in our 
2002 rule preamble which conveyed that our debottlenecking 
requirements would not change as a result of those rules and 
referred readers to a future rulemaking to address debottlenecking. 
This footnote has been read by some to suggest that debottlenecked 
units were required to continue to calculate emissions increases as 
they had under the prior rules. The intent of that footnote was not 
to express a position on how emissions increases were to be 
calculated at debottlenecked units but rather to make clear that the 
2002 NSR Improvement Rule would not change the fact that emissions 
from debottlenecked units must be included in the net emissions 
increase for the project, whenever appropriate, and that an upcoming 
rulemaking would, in accordance with the EPA recommendation to the 
President, address future treatment of debottlenecked units.

---------------------------------------------------------------------------

[[Page 54239]]

    As explained above, when an emissions increase is projected at a 
debottlenecked emissions unit, that increase must be added to the 
increase projected at the changed unit, along with the sum of all 
contemporaneous emissions increases and decreases, to determine whether 
NSR applies to the source. Consequently, even when a project increases 
emissions by less than a significant amount at the changed unit, the 
project would trigger major NSR if: (1) It debottlenecks another unit 
at the source; (2) the emissions increase \9\ (of that same pollutant) 
is large enough at the debottlenecked unit that there is a significant 
emissions increase resulting from the project; and (3) the 
contemporaneous emissions decreases and increases (of that same 
pollutant) at the source equal or exceed the levels that define a 
significant net emissions increase. If NSR applies, then the source 
goes through permitting, the changed unit undergoes a Best Available 
Control Technology (BACT) or Lowest Achievable Emissions Rate (LAER) 
analysis, and the net emissions increase is accounted for in the air 
quality analysis.
---------------------------------------------------------------------------

    \9\ As noted in footnote 4, later in this preamble we propose to 
include decreases from emissions units in calculating the emissions 
change that results from a project (i.e., in step 1 of the NSR 
applicability analysis).
---------------------------------------------------------------------------

B. Overview of This Proposed Action

    We propose to change the requirements for determining which 
emissions increases from existing units that are debottlenecked by a 
project elsewhere at the source must count towards NSR applicability. 
The purpose of this change is to remove barriers that the NSR program 
can impose that prevent owners and operators of major stationary 
sources from operating their facilities in the most efficient manner. 
Also, since 1992, EPA has worked to address concerns that the ``major 
NSR regulations were too complex and burdensome,'' \10\ and these 
proposed changes continue our efforts to simplify the process. Numerous 
commenters have previously identified debottlenecking changes as a 
particularly complex aspect of the NSR program. Among the improvements 
to NSR called for in the 2002 recommendations paper were changes to how 
these rules address debottlenecking of processes.
---------------------------------------------------------------------------

    \10\ See 61 FR 38250, 38252 (July 23, 1996).
---------------------------------------------------------------------------

    We propose to amend the relevant rules in light of not only our 
2002 energy policy recommendation for debottlenecking, but also 
consistent with the Court's holdings in New York. For purposes of 
clarity and greater certainty for affected parties, we propose that 
only those emissions increases at debottlenecked units that are 
``caused'' by the physical change or change in the method of operation 
be included in the modification analysis. We believe the 
debottlenecking regulations can be improved if, as described below, the 
causation requirement of the NSR rules is more appropriately tailored 
to circumstances where emissions increases clearly result from a 
proposed change.\11\ Our proposal seeks to refine the causation 
requirement, which we, in accordance with the D.C. Circuit ruling in 
New York, refer to as the ``but for'' causation requirement in light of 
various legal, physical and economic constraints that might exist on 
debottlenecked units. We are taking comment on all approaches to 
causation described below and ask whether it is more appropriate to 
rely on a single causation test or a combination of the tests.
---------------------------------------------------------------------------

    \11\ We intend for this rule to apply not only to emissions 
increases from debottlenecked units but also to any unchanged unit 
at a source that encounters an emissions increase after a project.
---------------------------------------------------------------------------

    As with our past policy, this debottlenecking rule proposal applies 
on a pollutant-specific basis. For example, a raw mill expansion at a 
cement plant is expected to result in a less than significant increase 
in particulate matter emissions. The increased raw mill capacity may 
also enable the previously constrained kiln to increase its productive 
capacity, thereby increasing emissions of other pollutants, such as 
nitrogen oxides (NOX). While there may not be a significant 
increase of particulate matter emissions from both units, there may be 
a significant increase of NOX emissions from the kiln. Since 
BACT or LAER cannot be triggered at a changed emissions unit unless the 
pollutant that has a significant net emissions increase is emitted by 
the changed unit, BACT or LAER would not apply to the raw mill 
expansion. PSD review, however, can be triggered for the source by 
increases in a pollutant not emitted by the changed unit.
    As noted above, we believe that it is appropriate to revisit the 
causation requirements for determining when an emissions increase at a 
debottlenecked unit is caused by a particular change elsewhere at the 
source. We do not believe that including emissions increases from 
debottlenecked units, without first establishing causation, is 
consistent with Congress' intent in establishing the major NSR program. 
As we explained in promulgating the demand growth exclusion, we 
interpret the ``which increases'' and ``which results in'' language of 
section 111(a)(4) of the modification provision of the CAA as requiring 
``a causal link between the proposed change and any post-change 
increase in emissions.'' See 67 FR 80203; New York, 413 F.3d at 32-33. 
In New York, the Court looked favorably on the demand growth exclusion 
for emissions increases that (1) Could have been accommodated prior to 
the change to meet the particular level of demand; and (2) were not 
caused by the change. See New York, 413 F.3d at 31-33.
    The EPA's interpretation of section 111(a)(4) as requiring a causal 
relationship is governed by Chevron U.S.A., Inc. v. Natural Res. Def. 
Council, 467 U.S. 837 (1984) (``Chevron''). This decision was explained 
in New York as follows:

As to EPA's interpretation of the CAA, we proceed under Chevron's 
familiar two-step process. See 467 U.S. at 842-43. In the first step 
(`Chevron Step 1'), we determine whether based on the Act's 
language, legislative history, structure, and purpose, `Congress has 
directly spoken to the precise question at issue.' Id. at 842. If 
so, EPA must obey. But if Congress's intent is ambiguous, we proceed 
to the second step (`Chevron Step 2') and consider `whether the 
agency's [interpretation] is based on a permissible construction of 
the statute.' Id. at 843. If so, we will give that interpretation 
`controlling weight unless [it is] arbitrary, capricious, or 
manifestly contrary to the statute.' Id. at 844.\12\

    \12\ See New York, 413 F.3d at 18.
---------------------------------------------------------------------------

    The EPA believes that even if Congress failed to articulate 
unambiguously that section 111(a)(4) requires a causal link between the 
proposed change and any post-change increase in emissions, the agency's 
approach is a reasonable interpretation of the statute and well within 
the purview of administrative deference under Chevron. Below, we 
describe various standards of causation that we believe are consistent 
with the statutory text of section 111(a)(4). The EPA believes that not 
only inferring causation under section 111(a)(4) would be entitled to 
deference, but that selection of one or more of these causation 
approaches would also be afforded similar deference under Chevron.

[[Page 54240]]

1. Legal Causation
    The causation test that is the most straightforward to apply and 
enforce for debottlenecked units would be a legal causation test in 
which an emissions increase at a debottlenecked unit would not be 
considered to have been caused by a physical or operational change at a 
major stationary source if the debottlenecked unit's post-project 
emissions were already authorized by a pre-existing air quality permit. 
This would apply to any debottlenecked unit with a permit that is 
enforceable as a practical matter.\13\ For example, if a unit is 
debottlenecked by a change elsewhere at the source, but it had 
previously been permitted (with a qualifying permit) to emit at 
operating levels that could be reached but would not be exceeded after 
the debottlenecking, under this legal causation test any change in 
emissions at this unit actually resulted from the initial authorization 
and not from the proposed change. The reasoning behind this 
interpretation is especially clear when considering units with 
nonattainment NSR permits, where the source must obtain offsets under 
its original permit for a level of emissions that represents the 
maximum operation allowed for the unit by its original permit. Thus, as 
long as the post-debottlenecking operational level is within their 
permitted limit, and the source has already obtained permanent offsets 
for operating at the higher level, it is logical to conclude that the 
change associated with the initial authorization ``caused'' the changed 
level of emissions.\14\
---------------------------------------------------------------------------

    \13\ ``Enforceable as a practical matter'' will be achieved if a 
requirement is both legally and practicably enforceable. A 
requirement is ``legally enforceable'' if some authority has the 
right to enforce the restriction. Under current EPA guidance, 
practicable enforceability for a source-specific permit will be 
achieved if the permit's provisions specify: (1) A technically 
accurate limitation and the portions of the source subject to the 
limitation; (2) the time period for the limitation (hourly, daily, 
monthly, and annual limits such as rolling annual limits); and (3) 
the method to determine compliance, including appropriate 
monitoring, recordkeeping, and reporting. For rules and general 
permits that apply to categories of sources, practicable 
enforceability additionally requires that the provisions: (1) 
Identify the types or categories of sources that are covered by the 
rule; (2) where coverage is optional, provide for notice to the 
permitting authority of the source's election to be covered by the 
rule; and (3) specify the enforcement consequences relevant to the 
rule.
    \14\ Here, we use nonattainment NSR as an example, but we 
propose to apply this approach to other types of air quality permits 
(i.e., PSD and Title V operating permits, and other permits that are 
enforceable as a practical matter).
---------------------------------------------------------------------------

    Under this test, the ``but for'' legal cause of the increase would 
be the original new construction or modification that received the 
initial emission authorization. Without this original event, and the 
accompanying permit, the emissions associated with debottlenecking 
could not have occurred. Accordingly, EPA believes it is the original 
event, not the debottlenecking event, that is the legal cause for the 
changes in emissions at the unchanged unit. Although it is possible 
hypothetically to attribute the emissions to either event, the presence 
of the permit is the basis on which to legally attribute the emissions 
to the event that gave rise to the permit limit. The emissions unit is 
legally constrained from operating at the post-change emissions rate, 
if such emissions would violate a legally and practically enforceable 
term or condition of any previously issued permit.
    The permit status of the unchanged unit would be the key criterion 
for establishing causation under this approach. For example, at a grey 
iron foundry, both the casters and rolling unit downstream of a melting 
tub are oversized; however, only the casting unit has a permit that is 
enforceable as a practical matter (e.g., Title V operating permit 
containing SIP limits). Due to a physical change to expand the capacity 
of the melting unit, the casting unit can operate at a higher 
throughput. If the casting unit has obtained a qualifying permit that 
authorized its higher operating level, the emissions associated with 
that operating level first achieved after the change at the melting 
unit would be legally caused by the change that resulted in the earlier 
permitting action (e.g., the original installation of the casting unit, 
or some modification to it) and not by the change at the melting 
unit.\15\ Conversely, for the rolling unit, which removes iron billet 
out of the caster, if it operates at higher levels after the change, 
but had not received authorization for its higher operating levels 
through a qualifying permit, we would not, under the legal causation 
approach, attribute the emissions increase to the original roller 
installation because there is no enforceable permit which serves as a 
basis for us to attribute the legal cause. Thus, the rolling unit's 
emissions increase--based on applying the actual-to-projected-actual 
test--would be attributable to the change and must be included in the 
overall emissions increase resulting from the expansion project at the 
melting unit.
---------------------------------------------------------------------------

    \15\ In the case where a casting unit emits at a level higher 
than its permitted emissions rate, then it is a change in the method 
of operation and may be subject to major NSR.
---------------------------------------------------------------------------

    We believe that this approach offers significant advantages to NSR 
implementation with virtually no disadvantage. The ``but for'' legal 
causation test would be simpler for owners and operators to apply and 
for reviewing authorities to administer. It would reduce the burden of 
complex source-wide emissions calculations that can involve disputes or 
confusion about the physical capabilities of the design of the 
unchanged unit absent the change elsewhere in the process. This burden 
and confusion would be eliminated where an existing permit already 
authorized the emissions increase. This approach also offers source 
owners and operators certainty in designing and planning projects at 
their sources, because they may rely on the air pollution decisions 
already made for a given unit when planning for the future operation of 
that unit. We further note that our current rules do not require BACT 
or LAER at unchanged units, so this policy would not result in less 
control on the unchanged unit. It may result in sources not needing 
BACT/LAER review for the changed units themselves in situations where 
the increase from the unchanged unit must be part of the NSR 
applicability calculation in order to reach significant increase levels 
for a pollutant emitted by the changed unit. However, in such cases, 
the emissions increase at the changed unit necessarily would have to be 
less than the de minimis significance levels, so any reduction in 
pollution would also be de minimis.
    While EPA believes that the legal causation approach may offer the 
greatest potential for improvement in the regulatory treatment of 
debottlenecking, we must address how this approach comports with the DC 
Circuit's recent decision in New York concerning Clean Units. The term 
``modification'' is defined by section 111(a)(4) as ``any physical 
change in, or change in the method of operation of, a stationary source 
which increases the amount of any air pollutant, emitted by such source 
or which results in the emission of any air pollutant not previously 
emitted.'' As previously stated, the agency has interpreted the ``which 
increases'' and ``which results in'' language of section 111(a)(4) as 
requiring a causal link between any change and any post-change increase 
in emissions. The EPA used this rationale in adopting the demand growth 
exclusion, and this exclusion was upheld by the court in New York. 
Therefore, under section 111(a)(4), there must be (1) A physical change 
or change in the method of operation, (2) that is the cause of, (3) an 
increase in emissions.

[[Page 54241]]

    In New York, the agency attempted to define Clean Unit status such 
that a change at the unit did not ``increase'' emissions for purposes 
of section 111(a)(4) as long as its status as a Clean Unit remained 
intact, even if the change caused an increase in actual emissions from 
the unit. See New York, 413 F.3d at 38. The court ruled that the agency 
lacked the authority to promulgate the Clean Unit provision because the 
term ``increases'' refers to an increase in actual emissions rather 
than potential or allowable emissions. This issue does not arise in 
this proposal, which focuses on the causation of the increase rather 
than its measurement.
    The agency believes that, with regard to debottlenecking, the CAA 
and section 111(a)(4) more specifically are silent as to what type of 
causation is required between the physical change or change in the 
method of operation and the increase in emissions that occurs at the 
debottlenecked unit. While the ``which increases'' and ``which results 
in'' language from section 111(a)(4) strongly suggests a causal 
relationship is required, the statutory text does not mandate nor 
offers explicit guidance concerning a specific approach or standard of 
causation. The EPA believes that the legal causation approach is a 
reasonable interpretation and construction of section 111(a)(4) and 
would therefore be subject to Chevron Step 2 deference afforded the 
agency in administering the NSR program.
    Under a legal causation test, we would view the original 
authorization of emissions from the unit to be the cause. If the 
emissions were authorized by a prior permit, then that prior 
transaction would be the cause of the emissions increase. If the 
emissions were not authorized previously, either because the permit 
level is exceeded or the unit failed to obtain a qualifying air quality 
permit, then the increase in emissions from the debottlenecked unit 
would be attributable not to a prior permit but instead to the change. 
Consistent with the Clean Unit portion of New York, we would count 
those emissions on an actual-to-projected-actual basis.
    The legal causation test addresses whether a change at one unit 
causes an emissions increase at another. This issue is distinct from 
the question addressed in the Clean Unit portion of New York, which 
focused on how to calculate the emissions of a changed unit where 
causation was not in question. A debottlenecked unit is not undergoing 
a change, so we must establish a basis for causation. The legal 
causation test uses as its basis the permit level authorized when the 
unit was previously permitted. If this level is exceeded, or if no 
prior permitting action authorized the emissions level that would 
define the basis, then the cause of those emissions would be the 
current change and an actual-to-projected-actual emissions analysis is 
required to determine the debottlenecked unit's emissions increase that 
is attributable to the change.
    Under this approach, an emissions increase at a debottlenecked 
emissions unit would be considered caused by the prior permitting 
action, and not by the project at issue, if the following three 
criteria are met:
     The unit's maximum emissions levels for each of the NSR 
pollutants in question is explicitly contained in a permit;
     The permit contains an allowable emissions limit (or 
operational limit that has the effect of constraining emissions) for 
the regulated NSR pollutant that is enforceable as a practical matter 
(e.g., Title V operating permit); and
     The unit itself is unchanged.\16\
---------------------------------------------------------------------------

    \16\ Under our existing regulations, exceeding a permit limit 
could be considered a change in the method of operation. Thus, while 
not physically changed, the debottlenecked unit would be 
operationally changed if it plans to exceed its prior permitted 
emissions limit.
---------------------------------------------------------------------------

    Under this legal causation approach for units meeting the above 
criteria, no future emissions increase at the debottlenecked unit is 
considered to have been caused by the project for the purposes of an 
NSR determination. In such circumstances, the contribution from the 
debottlenecked unit to determining whether the project results in a 
significant emissions increase is zero. On the other hand, if the 
project is expected to cause the debottlenecked emissions unit to 
increase above its permitted emissions, then its actual-to-projected-
actual emissions increase must be included in the emissions increase 
calculation. In addition, its underlying permit would require a change 
(i.e., to accommodate a higher permit limit), which would in most cases 
trigger review by the permitting authority.
    Under the legal causation test, the emissions increase from a 
proposed project involving a unit undergoing a physical or operational 
change and a debottlenecked unit is calculated as follows.
     For new units, the emissions increase equals the unit's 
potential to emit.
     For an existing emissions unit undergoing a physical 
change or change in the method of operation, the emissions increase is 
determined under the actual-to-projected-actual test \17\ as discussed 
in section III.A.3 of this preamble.
---------------------------------------------------------------------------

    \17\ States with approved programs may still require that 
sources use our past emissions increase test until their SIP 
revisions incorporating the 2002 rules are effective and approved.
---------------------------------------------------------------------------

     For a debottlenecked emissions unit that will not be 
changed and that is not subject to an emission limit that is 
enforceable as a practical matter, the emission increase is determined 
under the actual-to-projected-actual test.
     For a debottlenecked emissions unit that will not be 
changed and that is subject to an emission limit that is enforceable as 
a practical matter, the emissions increase is zero, unless the source 
plans to exceed its permitted level, in which case the emission 
increase is determined under the actual-to-projected-actual test.
     Add all of the emissions increases from the project as 
discussed above to determine whether there is a significant emissions 
increase as a result of the proposed project.\18\
---------------------------------------------------------------------------

    \18\ As noted in footnote 4, later in this preamble we propose 
to include decreases from emissions units in calculating the 
emissions change that results from a project (i.e., in step 1 of the 
NSR applicability analysis).
---------------------------------------------------------------------------

    Thus, all emissions increases that meet the causation test should 
be considered in the project's total emissions increase. This applies 
to all related units, even those that do not require a permit change 
after the project. Regardless of whether the related units require 
permit changes, under no circumstance can the source's new emissions 
level cause or contribute to a violation of the National Ambient Air 
Quality Standards (NAAQS) or an exceedance of the PSD increment. CAA 
Section 165(a)(3).
    It is important to note that the legal causation approach is not 
dependent on air quality modeling; rather, it is based on the fact that 
the reviewing authority has made an objective decision to authorize the 
unit to emit up to a certain level. Thus, we believe that a legal 
causation approach can effectively work with any unit that has a 
practically enforceable permit. This is because, as noted above, 
reviewing authorities have a statutory obligation to ensure that 
permitting in their jurisdictions will not cause or contribute to a 
violation of a NAAQS or PSD increment or adversely impact an air 
quality related value (AQRV) in a Class I area. Within each issued 
permit, even if it does not contain a comprehensive air quality 
assessment, the reviewing authority has responsibility for considering 
the totality of consequences of the source operating at the levels 
within the

[[Page 54242]]

permit. These consequences include, at a minimum, performing some 
screening of the local and regional impacts of the unit operating at 
the maximum allowed emissions level of the permit. The reviewing 
authority will make a determination based on, at a minimum, an air 
quality screening, emissions inventory review, or other means to ensure 
that the unit can operate up to that allowable limit and not violate 
the NAAQS or exceed the PSD increment. In making objective decisions, 
reviewing authorities must consider any public comment received. 
Accordingly, if the public is concerned about the air quality impacts 
related to a source's operation at a particular emissions level, and 
they raise specific, articulated concerns to the reviewing authority, 
the reviewing authority must address these concerns and ensure that no 
unacceptable, adverse impacts result from allowing the source to 
operate at the proposed new levels before issuing the permit.
    We solicit comment on all aspects of this preferred causation 
approach and on the proposed rule changes that implement this approach. 
We solicit comment on our proposal to apply legal causation to all 
permit limits that are enforceable as a practical matter (i.e., PSD and 
nonattainment permits, minor NSR permits, and other authorizations). We 
ask for comment on whether the legal causation approach may need to 
account for additional factors, as described in section III.C.3 of this 
preamble, such as the level of air quality or attainment modeling 
associated with the original permit limit. If so, how would it be 
appropriate to account for the factors? Should the legal causation 
approach be limited in application when the prior permit lacked air 
quality or attainment modeling?
2. Physical Causation
    A second approach to the causation requirement could focus upon a 
physical causation. Under this approach, the emissions increase at an 
unchanged unit would result from the change at the ``bottlenecking'' 
unit (and its emissions would be included in the project's emissions 
increase calculation) if the unchanged unit were physically incapable 
of operating at a higher level absent the change at the bottlenecking 
unit. An emissions unit is physically incapable of operating at the 
post-change emissions rate if pre-change operations at the major 
stationary source could not supply material to or accept material from 
the emissions unit due to inherent capacity constraints at the major 
stationary source, and there is no market from which or to which the 
major stationary source could purchase or sell the material, or if 
there is no other reasonable means of disposing of the material. In 
such a case, arguably the unchanged unit could not have physically 
accommodated its new emissions level but for the change.
    To help clarify a ``but for'' physical causation test, consider the 
example from above of the iron foundry that has oversized casters 
downstream of a smaller-sized melting tub. A physical change to expand 
the melting unit would allow the casting unit to operate at a higher 
throughput. ``But for'' the change at the melting unit, the iron 
casting unit could not have increased its operations because there 
could be no other physical supply of molten iron from any place other 
than the melting unit. Thus, increases in emissions from the 
debottlenecked casting unit that are attributable to the project at the 
melting tub would be assessed using an actual-to-projected-actual 
emissions test.
    In contrast, the ``but for'' physical causation would not exist in 
the case of the rolling unit at the foundry. The rolling unit is 
further downstream and removes iron billet out of the caster. However, 
the rolling unit could physically accommodate billet from other 
sources, since there is no physical impairment that would prevent the 
source from purchasing billet from other sources and increasing 
emissions from the rolling unit. Thus, a physical change at the casting 
unit (or further upstream, perhaps at the melting unit) would not be 
the ``but for'' cause of the emissions increase at the rolling unit and 
thus the rolling unit's emissions increase would not be attributable to 
the project.
    For another example, assume that the smelting of recycled aluminum 
at a secondary aluminum smelter and rolling mill is limited by the 
capacity of the smelter. The rolling mill, however, can produce product 
using aluminum ingots either from the secondary smelter or from a 
nearby primary aluminum plant. The source wants to expand the capacity 
of the smelter in order to utilize more recycled aluminum rather than 
buying ingots from the primary aluminum plant to meet its growing 
product demand. The rolling mill is not bottlenecked by the available 
smelter capacity since it can use, and has in the past used, other 
aluminum sources to produce its rolled aluminum products. In this case, 
the ``but for'' physical causation test is not met, and increases in 
emissions from the rolling mill would not be considered to be part of 
the project emissions. However, increases in emissions above its 
baseline emissions (highest 2 years in 10) would be contemporaneous 
emissions increases included in the netting analysis if the increase in 
smelter emissions were significant.
    We solicit comment on this approach in general. EPA anticipates 
that the emissions impacts of a physical causation approach are not 
appreciably different from those of a legal causation approach, but we 
anticipate that the improvements to certainty and clarity are 
diminished. Having to consider the physical capabilities of all 
emissions units at a source that are impacted by a project adds a 
degree of complexity to the causation evaluation. Whereas the ``but 
for'' legal causation boils down to whether or not the emissions 
increase was previously authorized, in this case there will need to be 
a technical judgment as to whether a source could have procured the 
input from another source. We solicit comment on how to most 
objectively determine what level an underutilized unit is physically 
capable of operating at, and, in general, how to most effectively 
evaluate projects using a but for physical causation test.
3. Economic Causation
    As an extension of the physical causation approach, we also take 
comment on whether causation should be tied to both physical and 
economic realities. Under this approach, emissions increases at 
debottlenecked units will not be considered to have resulted from the 
change if it would have been both physically possible and economically 
rational for the unchanged unit to have operated at the post-change 
level. Under this approach, in addition to those increases that result 
from physical causation as described above, an additional category of 
emissions increases would result from the change at the 
``bottlenecking'' unit (and their emissions would be included in the 
project's emissions increase calculation). This category would include 
units for which, although they may have been physically capable of 
operating at a higher level prior to the change at the bottlenecking 
unit, operating at the higher level would have been economically 
irrational. An emissions unit is economically constrained from 
operating at the post-change emissions rate, if a market exists from 
which or to which the major stationary source could purchase or sell 
the material, or if there is a reasonable means of disposing of the 
material, but the cost of such a transaction is so unreasonable it 
would preclude the major stationary source from engaging in the 
transaction.
    An example where a unit may have been able to physically 
accommodate

[[Page 54243]]

higher operating levels before a change to another unit but it would 
have been economically irrational to do so is an oversized boiler 
supplying steam to several pulp digesters at a paper mill. Conceivably, 
the boiler could have operated at higher capacity even though the 
digesters and all other parts of the mill were incapable of using the 
extra steam prior to making any change at any other part of the mill. 
The boiler could have simply blown off steam to maximize its emissions 
rate, and was physically (and possibly legally able to do so), but such 
a use of resources would only be to take advantage of regulatory 
provisions and would not otherwise serve an economically rational 
purpose. If the mill were to add more digesters and those digesters 
increased the demand for steam on the boiler, under this ``but for'' 
causation approach we would attribute the emissions increase to the 
physical change (i.e., adding the digesters) even though the boiler was 
physically capable of accommodating the increase prior to the change.
    While we are soliciting comment on the econom